Speaker Gaze Research Articles

The bigger your pupils, the better my comprehension: an ERP study of how pupil size and gaze of the speaker affect syntactic processing.

Gaze direction and pupil dilation play a critical role in communication and social interaction due to their ability to redirect and capture our attention and their relevance for emotional information. The present study aimed to explore whether the pupil size and gaze direction of the speaker affect language comprehension. Participants listened to sentences that could be correct or contain a syntactic anomaly, while the static face of a speaker was manipulated in terms of gaze direction (direct, averted) and pupil size (mydriasis, miosis). Left anterior negativity (LAN) and P600 linguistic event-related potential components were observed in response to syntactic anomalies across all conditions. The speaker's gaze did not impact syntactic comprehension. However, the amplitude of the LAN component for mydriasis (dilated pupil) was larger than for miosis (constricted pupil) condition. Larger pupils are generally associated with care, trust, interest, and attention, which might facilitate syntactic processing at early automatic stages. The result also supports the permeable and context-dependent nature of syntax. Previous studies also support an automatic nature of syntax (fast and efficient), which combined with the permeability to relevant sources of communicative information, such as pupil size and emotions, is highly adaptive for language comprehension and social interaction.

Developing a teacher agent to explore the role of gaze direction on young school children’s comprehension

Teachers’ gaze and school children’s language proficiency play a role for listening comprehension in the classroom. The influence of multimodal conditions on listening comprehension may be hard to study in a systematic manner with real speakers. The aim was to test the concept and feasibility of studying the influence of a speakers on passage comprehension performance in children using a digitally animated character. Twenty-four primary-school children were tested. A narrative passage was presented to the participants by a teacher agent. Half of the participants were randomized to a direct gaze condition and half to an averted gaze condition. The method is feasible with the intended population. We detected no differences in listening comprehension of the teacher agent’s gaze behavior in our sample. The concept works and it is feasible to test the effect of speaker gaze on children using the proposed methodology with some further development.

Does the speaker's eye gaze facilitate infants' word segmentation from continuous speech? An ERP study.

The environment in which infants learn language is multimodal and rich with social cues. Yet, the effects of such cues, such as eye contact, on early speech perception have not been closely examined. This study assessed the role of ostensive speech, signalled through the speaker's eye gaze direction, on infants' word segmentation abilities. A familiarisation-then-test paradigm was used while electroencephalography (EEG) was recorded. Ten-month-old Dutch-learning infants were familiarised with audio-visual stories in which a speaker recited four sentences with one repeated target word. The speaker addressed them either with direct or with averted gaze while speaking. In the test phase following each story, infants heard familiar and novel words presented via audio-only. Infants' familiarity with the words was assessed using event-related potentials (ERPs). As predicted, infants showed a negative-going ERP familiarity effect to the isolated familiarised words relative to the novel words over the left-frontal region of interest during the test phase. While the word familiarity effect did not differ as a function of the speaker's gaze over the left-frontal region of interest, there was also a (not predicted) positive-going early ERP familiarity effect over right fronto-central and central electrodes in the direct gaze condition only. This study provides electrophysiological evidence that infants can segment words from audio-visual speech, regardless of the ostensiveness of the speaker's communication. However, the speaker's gaze direction seems to influence the processing of familiar words. RESEARCH HIGHLIGHTS: We examined 10-month-old infants' ERP word familiarity response using audio-visual stories, in which a speaker addressed infants with direct or averted gaze while speaking. Ten-month-old infants can segment and recognise familiar words from audio-visual speech, indicated by their negative-going ERP response to familiar, relative to novel, words. This negative-going ERP word familiarity effect was present for isolated words over left-frontal electrodes regardless of whether the speaker offered eye contact while speaking. An additional positivity in response to familiar words was observed for direct gaze only, over right fronto-central and central electrodes.

The effects of referential gaze in spoken language comprehension: Human speaker vs. virtual agent listener gaze

IntroductionFour studies addressed effects of human speaker gaze vs. virtual agent listener gaze on eye movements during spoken sentence comprehension.MethodParticipants saw videos in which a static scene depicting three characters was presented on a screen. Eye movements were recorded as participants listened to German subject-verb-object (SVO) sentences describing an interaction between two of these characters. Participants' task was to verify whether the sentence matched a schematic depiction of the event. Two critical factors were manipulated across all four experiments: (1) whether the human speaker—uttering the sentence—was visible, and (2) whether the agent listener was present. Moreover, in Experiments 2 and 4, the target second noun phrase (NP2) was made inaudible, and in Experiments 3 and 4, the gaze time course of the agent listener was altered: it looked at the NP2 referent about 400 ms before the speaker did. These manipulations served to increase the value of the speaker's and listener's gaze cues for correctly anticipating the NP2 referent.ResultsHuman speaker gaze led to increased fixations of the NP2 referent in all experiments, but primarily after the onset of its mention. Only in Experiment 3 did participants reliably anticipate the NP2 referent, in this case making use of both the human speaker's and the virtual agent listener's gaze. In all other cases, virtual agent listener gaze had no effect on visual anticipation of the NP2 referent, even when it was the exclusive cue.DiscussionSuch information on the use of gaze cues can refine theoretical models of situated language processing and help to develop virtual agents that act as competent communication partners in conversations with human interlocutors.

When a look is enough: Neurophysiological correlates of referential speaker gaze in situated comprehension

Behavioral studies have shown that speaker gaze to objects in a co-present scene can influence listeners’ expectations about how the utterance will unfold. These findings have recently been supported by ERP studies that linked the underlying mechanisms of the integration of speaker gaze with an utterance meaning representation to multiple ERP components. This leads to the question, however, as to whether speaker gaze should be considered part of the communicative signal itself, such that the referential information conveyed by gaze can help listeners not only form expectations but also to confirm referential expectations induced by the prior linguistic context. In the current study, we investigated this question by conducting an ERP experiment (N=24, Age:[19,31]), in which referential expectations were established by linguistic context together with several depicted objects in the scene. Those expectations then could be confirmed by subsequent speaker gaze that preceded the referential expression. Participants were presented with a centrally positioned face performing gaze actions aligned to utterances comparing two out of three displayed objects, with the task to judge whether the sentence was true given the provided scene. We manipulated the gaze cue to be either Present (toward the subsequently named object) or Absent preceding contextually Expected or Unexpected referring nouns. The results provided strong evidence for gaze as being treated as an integral part of the communicative signal: While in the absence of gaze, effects of phonological verification (PMN), word meaning retrieval (N400) and sentence meaning integration/evaluation (P600) were found on the unexpected noun, in the presence of gaze effects of retrieval (N400) and integration/evaluation (P300) were solely found in response to the pre-referent gaze cue when it was directed toward the unexpected referent with attenuated effects on the following referring noun.

Turn-taking in human face-to-face interaction is multimodal: gaze direction and manual gestures aid the coordination of turn transitions.

Human communicative interaction is characterized by rapid and precise turn-taking. This is achieved by an intricate system that has been elucidated in the field of conversation analysis, based largely on the study of the auditory signal. This model suggests that transitions occur at points of possible completion identified in terms of linguistic units. Despite this, considerable evidence exists that visible bodily actions including gaze and gestures also play a role. To reconcile disparate models and observations in the literature, we combine qualitative and quantitative methods to analyse turn-taking in a corpus of multimodal interaction using eye-trackers and multiple cameras. We show that transitions seem to be inhibited when a speaker averts their gaze at a point of possible turn completion, or when a speaker produces gestures which are beginning or unfinished at such points. We further show that while the direction of a speaker's gaze does not affect the speed of transitions, the production of manual gestures does: turns with gestures have faster transitions. Our findings suggest that the coordination of transitions involves not only linguistic resources but also visual gestural ones and that the transition-relevance places in turns are multimodal in nature. This article is part of a discussion meeting issue 'Face2face: advancing the science of social interaction'.

Generation Model for Head Nods Consistent with Features of Verbal Response Tokens

In human-human interactions, a listener uses both verbal tokens and head nods for responding signals, and they frequently co-occur. When humanoid robots and anthropomorphic agents response to a user using verbal tokens and head nods simultaneously, they must be generated in proper timing to each other and have consistent features. In this paper, we propose models to predict co-occurrence and physical features of head nods based on prosodic and syntactic features of verbal response tokens. We used, as predictive variables, the forms, positions, durations, averages/standard deviations of fundamental frequency and loudness of response tokens and head positions at the beginning of response tokens. In addition, considering participation framework, we also used speaker's gaze and listener's gaze at the beginning of response tokens, and applied generalized mixed models to predict the co-occurrence, type, range, repetition and velocity of head nods. The results confirmed that proposed models can predict these outcomes effectively.

When eyes beat lips: speaker gaze affects audiovisual integration in the McGurk illusion

Eye contact is a dynamic social signal that captures attention and plays a critical role in human communication. In particular, direct gaze often accompanies communicative acts in an ostensive function: a speaker directs her gaze towards the addressee to highlight the fact that this message is being intentionally communicated to her. The addressee, in turn, integrates the speaker’s auditory and visual speech signals (i.e., her vocal sounds and lip movements) into a unitary percept. It is an open question whether the speaker’s gaze affects how the addressee integrates the speaker’s multisensory speech signals. We investigated this question using the classic McGurk illusion, an illusory percept created by presenting mismatching auditory (vocal sounds) and visual information (speaker’s lip movements). Specifically, we manipulated whether the speaker (a) moved his eyelids up/down (i.e., open/closed his eyes) prior to speaking or did not show any eye motion, and (b) spoke with open or closed eyes. When the speaker’s eyes moved (i.e., opened or closed) before an utterance, and when the speaker spoke with closed eyes, the McGurk illusion was weakened (i.e., addressees reported significantly fewer illusory percepts). In line with previous research, this suggests that motion (opening or closing), as well as the closed state of the speaker’s eyes, captured addressees’ attention, thereby reducing the influence of the speaker’s lip movements on the addressees’ audiovisual integration process. Our findings reaffirm the power of speaker gaze to guide attention, showing that its dynamics can modulate low-level processes such as the integration of multisensory speech signals.

Is Conversation Built for Two? The Partitioning of Social Interaction

ABSTRACT Conversation is flexible enough to be conducted with varying numbers of individuals, but most conversation is dyadic. Is the prevalence of dyadic focal participation frameworks facilitated by structures of conversation? Using video recordings of spontaneous naturally occurring conversations, I explore multiperson interactions focusing on how structures of turn taking, sequence organization, storytelling, and speaker gaze facilitate or inhibit the inclusion of multiple individuals in conversation. As I show, because our system favors dyadic participation through turn allocation and sequence organization, sustaining focal triadic or multiparty participation frameworks requires more interactional work than sustaining a dyadic focal participation framework. However, serially dyadic participation, which keeps dyads shifting, and story- and joke telling facilitate the participation of multiple individuals. Although conversational structures can be adapted to partition focal participation as dyadic or multiparty on a moment-by-moment basis, the structures generally facilitate dyadic focal participation. Data are in American English.

Word learning in monolingual and bilingual children: The influence of speaker eye-gaze.

The current study examined the impact of a speaker's gaze on novel-word learning in 4-5-year old monolingual (N = 23) and bilingual children (N = 24). Children were taught novel words when the speaker looked at the object both times while labeling it (consistent) and when the speaker looked at the object only the first time (inconsistent). During teaching, bilingual children differentiated between the target object (that matched the label) and non-target object (that did not match the label) earlier than the monolingual children on trials without eye-gaze information. However, during testing, monolingual children showed more robust retention of novel words than bilingual children in both conditions. Findings suggest that bilingualism shapes children's attention to eye-gaze during word learning, but that, ultimately, there is no bilingual advantage for utilizing this cue in the service of word retention.

The role of speaker eye gaze and mutual exclusivity in novel word learning by monolingual and bilingual children

The current study examined the combined effect of a speaker’s eye gaze and mutual exclusivity (ME) on novel word retention in monolingual and bilingual children. A novel object was presented with a familiar object, and children were taught new labels for objects under two conditions. In the Align condition, the speaker’s gaze and the ME cue provided the same information (the speaker looked at the novel object while labeling it with a novel name). In the Conflict condition, the speaker’s gaze and the ME cue provided competing information (the speaker looked at the familiar object while labeling it with a novel name). Using a visual world eye-tracking paradigm, children’s retention was assessed by testing novel objects with novel labels and by testing the familiar objects with novel labels. We found that all children successfully retained the novel labels for novel objects when both eye gaze and ME provided the same information. However, when the cues conflicted, bilingual children did not perform above chance for either novel objects or familiar objects. In contrast, monolingual children demonstrated retention of novel labels for familiar objects but not for novel objects. Together, the findings suggest that redundant cues benefit word retention in all children regardless of linguistic background. Furthermore, when speaker gaze and ME conflict, bilingual children appear to disregard both cues during retention, whereas monolingual children may be more willing to retain novel labels for familiar words, suggesting that they prioritize a speaker’s eye gaze over ME.

Influence of speakers’ gaze on situated language comprehension: Evidence from Event-Related Potentials

Behavioral studies have shown that speaker gaze to objects in a co-present scene can influence listeners’ sentence comprehension. To gain deeper insight into the mechanisms involved in gaze processing and integration, we conducted two ERP experiments (N = 30, Age: [18, 32] and [19, 33] respectively). Participants watched a centrally positioned face performing gaze actions aligned to utterances comparing two out of three displayed objects. They were asked to judge whether the sentence was true given the provided scene. We manipulated the second gaze cue to be either Congruent (baseline), Incongruent or Averted (Exp1)/Mutual (Exp2). When speaker gaze is used to form lexical expectations about upcoming referents, we found an attenuated N200 when phonological information confirms these expectations (Congruent). Similarly, we observed attenuated N400 amplitudes when gaze-cued expectations (Congruent) facilitate lexical retrieval. Crucially, only a violation of gaze-cued lexical expectations (Incongruent) leads to a P600 effect, suggesting the necessity to revise the mental representation of the situation. Our results support the hypothesis that gaze is utilized above and beyond simply enhancing a cued object’s prominence. Rather, gaze to objects leads to their integration into the mental representation of the situation before they are mentioned.

Demystifying infant vocal imitation: The roles of mouth looking and speaker's gaze.

Vocal imitation plays a fundamental role in human language acquisition from infancy. Little is known, however, about how infants imitate other's sounds. We focused on three factors: (a) whether infants receive information from upright faces, (b) the infant's observation of the speaker's mouth and (c) the speaker directing their gaze towards the infant. We recorded the eye movements of 6-month-olds who participated in experiments watching videos of a speaker producing vowel sounds. We found that an infants' tendency to vocally imitate such videos increased as a function of (a) seeing upright rather than inverted faces, (b) their increased looking towards the speaker's mouth and (c) whether the speaker directed their gaze towards, rather than away from infants. These latter findings are consistent with theories of motor resonance and natural pedagogy respectively. New light has been shed on the cues and underlying mechanisms linking infant speech perception and production.

Embodying epistemic responsibility

The study explores how children deploy gaze and embodied epistemic stance displays to establish a mutual epistemic responsibility when dealing with potentially controversial questions. Drawing on video recordings of 24 peer interactions involving children aged 9-12 years, the sequential and multimodal analysis describes the practices that construct intercorporeal participation frameworks for collaborative reasoning. Findings demonstrate that children coordinate gaze and multimodal displays of epistemic stance to mobilize co-participants' attention toward their position, while at the same time subjecting it to negotiation. Furthermore, children recruit the current speaker's gaze to issue a friendly challenge to his/her pre-determined stance. When the mutual epistemic responsibility was at stake, children occasioned a recalibration of stance displays at the earliest possible place. The children's embodied participation frameworks thus reflect their orientation to knowledge as being socially constructed.

Eye Tracking During Visually Situated Language Comprehension: Flexibility and Limitations in Uncovering Visual Context Effects.

The present work is a description and an assessment of a methodology designed to quantify different aspects of the interaction between language processing and the perception of the visual world. The recording of eye-gaze patterns has provided good evidence for the contribution of both the visual context and linguistic/world knowledge to language comprehension. Initial research assessed object-context effects to testtheories of modularity in language processing. In the introduction, we describe how subsequent investigations have taken the role of the wider visual context in language processing as a research topic in its own right, asking questions such as how our visual perception of events and of speakers contributes tocomprehension informed bycomprehenders' experience. Among the examined aspects of the visual context are actions, events, a speaker's gaze, and emotional facial expressions, as well as spatial object configurations. Following an overview of the eye-tracking method and its different applications, we list the key steps of the methodology in the protocol, illustrating how to successfully use it to study visually-situated language comprehension. A final section presents three sets of representative results and illustrates the benefits and limitations of eye tracking for investigating the interplay between the perception of the visual world and language comprehension.

Distinct effects of different visual cues on sentence comprehension and later recall: The case of speaker gaze versus depicted actions

Language-processing accounts are beginning to accommodate different visual context effects, but they remain underspecified regarding differences between cues, both during sentence comprehension and subsequent recall. We monitored participants' eye movements to mentioned characters while they listened to transitive sentences. We varied whether speaker gaze, a depicted action, neither, or both of these visual cues were available, as well as whether both cues were deictic (Experiment 1) or only speaker gaze (Experiment 2). Speaker gaze affected eye movements during comprehension similarly early to a single deictic action depiction, but significantly earlier than non-deictic action depictions; conversely, depicted actions but not speaker gaze positively affected later recall of sentence content. Thus, cue type and cue-language relations must be accommodated in characterising real-time situated language comprehension and subsequent recall of sentence content.

(Simulated) listener gaze in real‐time spoken interaction

AbstractGaze is an important aspect of social communication. Previous research has concentrated mainly on the role of speaker gaze and listener gaze in isolation, neglecting the effect of the listener's gaze behavior on the speaker's behavior. This paper presents an exploratory eye‐tracking study involving an interactive human‐like agent following participants' gaze. This study demonstrates that a rather simple gaze‐following mechanism convincingly simulates active listening behavior engaging the speaker. The study also highlights how speakers rely on their interlocutors' gaze when establishing common references.

Learning to Predict Readability Using Eye-Movement Data From Natives and Learners

Readability assessment can improve the quality of assisting technologies aimed at language learners. Eye-tracking data has been used for both inducing and evaluating general-purpose NLP/AI models, and below we show that unsurprisingly, gaze data from language learners can also improve multi-task readability assessment models. This is unsurprising, since the gaze data records the reading difficulties ofthe learners. Unfortunately, eye-tracking data from language learners is often much harder to obtain than eye-tracking data from native speakers. We therefore compare the performance of deep learning readability models that use nativespeaker eye movement data to models using data from language learners. Somewhat surprisingly, we observe no significant drop in performance when replacing learners with natives, making approaches that rely on native speaker gaze information, more scalable. In other words, our finding is that language learner difficulties can be efficiently estimated from native speakers, which suggests that, more generally, readily available gaze data can be used to improve educational NLP/AI models targeted towards language learners.

Does dynamic information about the speaker's face contribute to semantic speech processing? ERP evidence

Face-to-face interactions characterize communication in social contexts. These situations are typically multimodal, requiring the integration of linguistic auditory input with facial information from the speaker. In particular, eye gaze and visual speech provide the listener with social and linguistic information, respectively. Despite the importance of this context for an ecological study of language, research on audiovisual integration has mainly focused on the phonological level, leaving aside effects on semantic comprehension. Here we used event-related potentials (ERPs) to investigate the influence of facial dynamic information on semantic processing of connected speech. Participants were presented with either a video or a still picture of the speaker, concomitant to auditory sentences. Along three experiments, we manipulated the presence or absence of the speaker's dynamic facial features (mouth and eyes) and compared the amplitudes of the semantic N400 elicited by unexpected words. Contrary to our predictions, the N400 was not modulated by dynamic facial information; therefore, semantic processing seems to be unaffected by the speaker's gaze and visual speech. Even though, during the processing of expected words, dynamic faces elicited a long-lasting late posterior positivity compared to the static condition. This effect was significantly reduced when the mouth of the speaker was covered. Our findings may indicate an increase of attentional processing to richer communicative contexts. The present findings also demonstrate that in natural communicative face-to-face encounters, perceiving the face of a speaker in motion provides supplementary information that is taken into account by the listener, especially when auditory comprehension is non-demanding.

Foreground gesture, background gesture

Abstract Do speakers intend their gestures to communicate? Central as this question is to the study of gesture, researchers cannot seem to agree on the answer. According to one common framing, gestures are an “unwitting” window into the mind (McNeill, 1992); but, according to another common framing, they are designed along with speech to form “composite utterances” (Enfield, 2009). These two framings correspond to two cultures within gesture studies – the first cognitive and the second interactive in orientation – and they appear to make incompatible claims. In this article I attempt to bridge the cultures by developing a distinction between foreground gestures and background gestures. Foreground gestures are designed in their particulars to communicate a critical part of the speaker’s message; background gestures are not designed in this way. These are two fundamentally different kinds of gesture, not two different ways of framing the same monolithic behavior. Foreground gestures can often be identified by one or more of the following hallmarks: they are produced along with demonstratives; they are produced in the absence of speech; they are co-organized with speaker gaze; and they are produced with conspicuous effort. The distinction between foreground and background gestures helps dissolve the apparent tension between the two cultures: interactional researchers have focused on foreground gestures and elevated them to the status of a prototype, whereas cognitive researchers have done the same with background gestures. The distinction also generates a number of testable predictions about gesture production and understanding, and it opens up new lines of inquiry into gesture across child development and across cultures.